KRAS, to a lesser extent NRAS, and rarely HRAS, are mutated in a third of human cancers, which is well established to promote tumorigenesis. Using two independent approaches to manipulate oncogenic KRas signaling at the endogenous level, we show that the magnitude of this signaling causes opposite effects depending on the stage of tumorigenesis. Specifically, we reported in the first grant submission that wild-type HRas and NRas proteins are activated downstream of oncogenic KRas, and when knocked down, inhibited the tumor growth of cancer cell lines. Conversely, we now find that genetic ablation of wild-type Hras1 in mice promotes early oncogenic KRas-driven tumorigenesis. Independent of this line of investigation, we recently discovered that KRas is expressed very poorly compared to HRas owing to a bias of rare codons. Changing the rare to common codons produced more oncogenic KRas protein, enhanced tumor growth of established cell lines, but inhibited development of early lesions in mice. Since oncogenic Ras can induce a senescent growth arrest in normal cells, we hypothesize that KRas is expressed in a very narrow range in normal cells - high enough to promote proliferation but low enough to avoid excessive senescence - leading to hyperplasia when mutated. We propose to test this hypothesis in AIM 1 by determining the impact of manipulating the expression of wild-type Ras proteins and the codon bias of KRAS on senescence during early tumorigenesis. However, low oncogenic KRas signaling nevertheless inhibited tumor growth of established cell lines. As such, we hypothesize that there is selective pressure to overcome poor KRas expression imposed by rare codons during later stages of tumor development. In this regard, there is a gain in the copy number of the mutant KRAS allele in many human cancers, which is associated with a worse prognosis. We thus propose in AIM 2 to test whether changing the rare to common codons in settings that tolerate high oncogene expression suppresses a gain in Kras copy number during tumorigenesis. Our preliminary findings indicate that a gain in copy number is not the only mechanism to increase KRas protein expression. Thus, we also propose to identify how else KRas protein expression is elevated during tumorigenesis, and whether codon bias also regulates the expression of other cancer-related genes. Completion of this study will elucidate the molecular mechanisms underlying how the level of endogenous oncogenic KRas signaling differentially impacts early and late tumorigenesis.